WASHINGTON (NYTimes) - Concerned that the missile defence system designed to protect US cities is insufficient by itself to deter a North Korean attack, the Trump administration is expanding its strategy to also try to stop Pyongyang's missiles before they get far from Korean airspace.
The new approach, hinted at in an emergency request to Congress last week for US$4 billion to deal with North Korea, envisions the stepped-up use of cyber weapons to interfere with the North's control systems before missiles are launched, as well as drones and fighter jets to shoot them down moments after liftoff.
The missile defence network on the West Coast would be expanded for use if everything else fails.
In interviews, defence officials, along with top scientists and senior members of Congress, described the accelerated effort as a response to the unexpected progress that North Korea has made in developing intercontinental ballistic missiles capable of delivering a nuclear weapon to the continental United States.
"It is an all-out effort," said Senator Jack Reed of Rhode Island, the top Democrat on the Senate Armed Services Committee, who returned from a lengthy visit to South Korea last month convinced that the United States needed to do far more to counter North Korea.
"There is a fast-emerging threat, a diminishing window, and a recognition that we can't be reliant on one solution."
For years, that single solution has been the missile batteries in Alaska and California that would target any long-range warheads fired toward the US mainland, trying to shoot them down as they re-enter the atmosphere. Such an approach, known as "hitting a bullet with a bullet", remains of dubious effectiveness, even after more than $100 billion has been spent on the effort. Anti-missile batteries on ships off the Korean coast and in South Korea protect against medium-range missiles but not those aimed at US cities.
So, the administration plans to pour hundreds of millions of dollars into the two other approaches, both of which are still in the experimental stage.
The first involves stepped-up cyberattacks and other sabotage that would interfere with missile launches before they occur - what the Pentagon calls "left of launch".
The second is a new approach to blowing up the missiles in the "boost phase", when they are slow-moving, highly visible targets.
President Donald Trump has praised the existing missile defence system, insisting last month that it "can knock out a missile in the air 97 per cent of the time", a claim that arms control experts call patently false.
In trial runs, conducted under ideal conditions, the interceptors in Alaska and California have failed half of the time. And the Pentagon has warned administration officials that the North will soon have enough long-range missiles to launch volleys of them, including decoys, making the problem far more complex.
That helps explain the rush for new protections.
"They're looking at everything," said Thomas Karako, a senior fellow at the Center for Strategic and International Studies in Washington, who recently led two anti-missile studies and closely monitors the administration's planning.
"What you're seeing is a lot more options on the table."
The US$4 billion emergency budget sought by the White House is on top of the US$8 billion that the Missile Defence Agency has already been granted for this fiscal year, as well as what other military services and agencies are putting into missile defence.
Another US$440 million was moved from existing programs to anti-missile work two months ago, as the North Korea threat became more serious.
In the emergency request to Congress, and in documents made public by its committees, the precise use of the funds is cloaked in deliberately vague language.
Hundreds of millions of dollars, for example, are allotted for what the documents called "disruption/defeat" efforts.
Several officials confirmed that the "disruption" efforts include another, more sophisticated attempt at the kind of cyber and electronic strikes that former president Barack Obama ordered in 2014 when he intensified his efforts to cripple North Korea's missile testing.
Using cyber weapons to disrupt launches is a radical innovation in missile defence in the past three decades.
But in the case of North Korea, it is also the most difficult. It requires getting into the missile manufacturing, launch control and guidance systems of a country that makes very limited use of the internet and has few connections to the outside world - most of them through China and, to a lesser degree, Russia.
In the operation that began in 2014, a range of cyber and electronic-interference operations were used against the North's Musudan intermediate-range missiles, in an effort to slow its testing. But that secret effort had mixed results.
The failure rate for the Musudan missile soared to 88 per cent, but it was never clear how much of that was due to the cyberattacks and how much to sabotage of the North's supply chain and its own manufacturing errors. Then Kim Jong Un, the country's President, ordered a change in design, and the test-launches have been far more successful.
The experience has raised difficult questions about the effectiveness of cyber weapons, despite billions of dollars in investment.
"We can dream of a lot of targets to hack," said Michael Sulmeyer, director of the Cyber Security Project at Harvard and formerly the director for cyberpolicy planning and operations in the office of the secretary of defence.
"But it can be hard to achieve the effects we want, when we want them."
Congressional documents also talk of making "additional investments" in "boost-phase missile defence".
The goal of that approach is to hit long-range missiles at their point of greatest vulnerability - while their engines are firing and the vehicles are stressed to the breaking point, and before their warheads are deployed.
Defence Secretary Jim Mattis is also weighing, among other boost-phase plans, formulas that draw on existing technologies and could be deployed quickly.
One idea is having stealth fighters such as the F-22 or the F-35 scramble from nearby bases in South Korea and Japan at the first sign of North Korean launch preparations. The jets would carry conventional air-to-air missiles, which are 12 feet long, and fire them at the North Korean long-range missiles after they are launched.
But they would have to fly relatively close to North Korea to do that, increasing the chances of being shot down.
A drawback of boost-phase defence is the short window to use it. Long-range missiles fire their engines for just five minutes or so, in contrast to warheads that zip through space for about 20 minutes before plunging back to Earth. And there is the risk of inviting retaliation from North Korea.
"You have to make a decision to fire a weapon into somebody's territory," General John E. Hyten of the Air Force, commander of the US Strategic Command, which controls the US nuclear missile fleet, recently told a Washington group. "And if you're wrong, or if you miss?"
A boost-phase idea getting much notice would be to have drones patrol high over the Sea of Japan, awaiting a North Korean launch. Remote operators would fire heat-sensing rockets that lock onto the rising missiles.
"It's a huge advance," Gerold Yonas, chief scientist for former president Ronald Reagan's "Star Wars" program, said of the drone plan.
"It's one of those things where you hit yourself on the forehead and say, 'Why didn't I think of that?'"
Leonard H. Caveny, a main planner of the rocket-firing drones and a former Navy officer who directed science and technology at the Pentagon's anti-missile program from 1985 to 1997, said an accelerated programme could produce the weapons in a year or less.
Caveny's team is considering use of the Avenger, a drone made by General Atomics that has a wingspan of 76 feet.
"This is going to be a game changer," said Arthur L. Herman, a senior fellow at the Hudson Institute in Washington, who collaborates with Caveny.
The Pentagon's Missile Defence Agency is also developing a drone that would fire potent laser beams at rising missiles.
But recent plans would have it make its debut no sooner than 2025 - too late to play a role in the current crisis or the Trump presidency.
Even so, the effort has influential backers. In the recent talk, Hyten of Strategic Command called lasers much better than interceptor rockets because they avoided questions over firing weapons into sovereign territories, especially to knock out missile test-flights.
A potent beam of highly concentrated light, he said, "goes out into space", avoiding the trespassing issue.
In recent months, Congress has urged Pentagon officials to develop both varieties of drones.
Theodore A. Postol, a professor emeritus of science and national security policy at the Massachusetts Institute of Technology who has drawn up plans for a missile-firing drone, argued that fleets of such weapons patrolling near the North, threatening to undo its strategic forces, would be extremely intimidating and create new diplomatic leverage.
"We need it now," he said. "My concern is that we get something out there quickly that will pressure North Korea to negotiate."